kentcdodds

kentcdodds / kody

Public

An experimental personal assistant platform built on Cloudflare Workers and MCP

18
1
100% credibility
Found Mar 28, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Kody is an experimental personal AI assistant for controlling smart home devices like Roku TVs, Lutron lights, and Samsung TVs through natural chat.

How It Works

1
🔍 Discover Kody

You hear about a friendly personal AI helper that chats with you to control your smart home devices like TVs and lights.

2
🚀 Get started easily

Run one simple command to set up your own assistant on your computer.

3
⚙️ Launch your helper

Start the servers and open your web browser to see your assistant ready to go.

4
🏠 Connect your devices

Your helper scans and links to your Roku, lights, and TV automatically on your home network.

5
💬 Chat and command

Talk naturally to your AI like 'turn on the living room lights' and watch it happen.

✨ Home runs smoothly

Enjoy your smart home responding effortlessly to your voice through your personal AI buddy.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is kody?

Kody is an experimental personal assistant platform built with TypeScript on Cloudflare Workers and Remix, designed for single-user home automation. You chat via AI agents supporting MCP—like ChatGPT—to control Roku TVs, Lutron lights, and Samsung TVs through a local home connector that handles discovery, auth, and commands. Quick start with `npx create-epicflare` spins up a dev server with mocks, OAuth UI for secrets, and endpoints like `/health` or device status pages.

Why is it gaining traction?

Its MCP-first design keeps tools compact—a search endpoint plus code execution—avoiding bloated catalogs, while working across agent hosts. Dev perks include CLI restarts (`r` key), auto-mocking for AI/GitHub/Cloudflare APIs, and E2E tests via Playwright. Kent C. Dodds' experimental GitHub project draws tinkerers eyeing github experimental features for personal control.

Who should use this?

Home automation hackers building AI-driven setups for lights, TVs, and shades. Cloudflare devs prototyping agent tools without multi-tenant hassle. Solo users like podcasters tweaking studio lights via voice, not frontend teams.

Verdict

Promising prototype (18 stars, 1.0% credibility score) with strong tests and setup docs—great for experimental personality research in AI home control. Fork for personal jets or essays on smart homes, but skip production until multi-user support lands.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.